Color Measurement of Tea Leaves at
Different Drying Periods Using
Hyperspectral Imaging Technique
Chuanqi Xie1,2, Xiaoli Li1, Yongni Shao1*, Yong He1,3*
1.College of Biosystems Engineering and Food Science, Zhejiang University, Hangzhou, China,2.
Agricultural and Biological Engineering Department, University of Florida, Gainesville, Florida, United States of America,3.Key Laboratory of Equipment and Informatization in Environment Controlled Agriculture, Ministry of Agriculture, Hangzhou, China
*ynshao@zju.edu.cn (YS);yhe@zju.edu.cn (YH)
Abstract
This study investigated the feasibility of using hyperspectral imaging technique for nondestructive measurement of color components (DL*,Da* andDb*) and classify tea leaves during different drying periods. Hyperspectral images of tea leaves at five drying periods were acquired in the spectral region of 380–1030 nm. The three color features were measured by the colorimeter. Different preprocessing
algorithms were applied to select the best one in accordance with the prediction results of partial least squares regression (PLSR) models. Competitive adaptive reweighted sampling (CARS) and successive projections algorithm (SPA) were used to identify the effective wavelengths, respectively. Different models (least squares-support vector machine [LS-SVM], PLSR, principal components
regression [PCR] and multiple linear regression [MLR]) were established to predict the three color components, respectively. SPA-LS-SVM model performed
excellently with the correlation coefficient (rp) of 0.929 for DL*, 0.849 for Da*and
0.917 for Db*, respectively. LS-SVM model was built for the classification of different tea leaves. The correct classification rates (CCRs) ranged from 89.29% to 100% in the calibration set and from 71.43% to 100% in the prediction set, respectively. The total classification results were 96.43% in the calibration set and 85.71% in the prediction set. The result showed that hyperspectral imaging technique could be used as an objective and nondestructive method to determine color features and classify tea leaves at different drying periods.
OPEN ACCESS
Citation:Xie C, Li X, Shao Y, He Y (2014) Color Measurement of Tea Leaves at Different Drying Periods Using Hyperspectral Imaging Technique. PLoS ONE 9(12): e113422. doi:10. 1371/journal.pone.0113422
Editor:Ritaban Dutta, CSIRO, Australia
Received:July 25, 2014
Accepted:October 23, 2014
Published:December 29, 2014
Copyright:ß2014 Xie et al. This is an
open-access article distributed under the terms of the
Creative Commons Attribution License, which permits unrestricted use, distribution, and repro-duction in any medium, provided the original author and source are credited.
Data Availability:The authors confirm that all data underlying the findings are fully available without restriction. All relevant data are within the paper.
Funding:This work was supported by 863 National High-Tech Research and Development Plan (2013AA102301, 2011AA100705), Zhejiang Province Department of education research project (Y201327409), the Fundamental Research Funds for the Central Universities of China
(2012FZA6005, 2013QNA6011) and National Natural Science Foundation of China (61201073).
Introduction
Tea is welcome by many people because of its healthy function. For example, it can prevent cancer and cardiovascular disease and cure chronic gastritis [1], [2]. Tea processing procedure, which is composed of a series of physical and chemical reactions, can affect tea’s quality directly [3]. However, the change of color values (DL*,Da* andDb*) of tea leaves play significant roles in tea processing procedure. Therefore, studying color parameters of tea leaves during drying periods can finally improve tea’s quality.
Hyperspectral imaging technique, which integrates both spectral and imaging techniques, has been widely applied in many fields [4], [5], [6]. A spatial picture can be generated when the sample is scanned by the hyperspectral imaging system. The spatial picture (hyperspectral cube) consists of a series of images at different wavelength, and each pixel has both spectroscopic and spatial information. The schematic hyperspectral imaging system can be seen in Fig. 1. In accordance with the previous studies, hyperspectral imaging technique is very efficient for knowing the process when the samples changes with time [7], [8].
Hyperspectral technique has many advantages, such as nondestructive, rapid and simple operation, accurate, low cost, and can be applied in on-line detection. At present, hyperspectral technique has already been used to detect color
parameters in many studies [9], [10], [11], [12], [13]. The change of color parameters of tea leaves is very important in the tea processing procedure. However, to measure the color parameters of tea leaves at different drying periods using hyperspectral imaging technique has not been found.
The goals of this work were: (1) to find the quantitative relationships between the spectral reflectance information and color parameters of tea leaves at different drying periods; (2) to obtain effective wavelengths which are useful for the determination of color values; (3) to compare the predictive ability of different calibration models; (4) to develop an algorithm for the determination of color values of tea leaves.
Materials and Methods
Hyperspectral imaging system
A visible and near infrared (VIS-NIR) hyperspectral imaging system covering the spectral wavelengths of 380–1030 nm was used in this study (as shown inFig. 2). The system includes a lens (OLE-23), an imaging spectrograph (V10E-QE, Specim, Finland), a CCD camera (C8484-05, Hamamatsu City, Japan), two light sources (Oriel Instruments, Irvine, USA) provided by two 150W quartz tungsten halogen lamps, a conveyer belt operated by a stepper motor (IRCP0076, Isuzu Optics Corp., Taiwan, China), and a computer operating the spectral image system V10E software (Isuzu Optics Corp., Taiwan, China). The area CCD array detector of the camera has 6726512 (spatial6 spectral) pixels, and the spectral
Fig. 1. Hyperspectral imaging.
detector in spatial-spectral(x6l) axes. The ENVI 4.7 (Research system Inc.,
Boulder, Co., USA), Unscrambler V9.7 (CAMO Process AS, Oslo, Norway) and Matlab R2009a (The Math Works, Inc., Natick, MA, USA) software were used in this study.
Samples preparation and flow chart
Three different cultivars of tea leaves including Biyun, Longjing-43 and
Zhongcha-302 were used in this study. All of them were very famous tea in China. The number of each cultivar was fourteen. All leaves were picked from the green house, which is located at Zhejiang University, Hangzhou (120.2E, 30.3N), China. First of all, hyperspectral images of forty-two fresh tea leaves were acquired, and then three color parameters (DL*,Da* and Db*) of these fresh tea leaves were measured using the colorimeter (Konica Minolta, SC-80C, Japan). These leaves were then dried in a drying oven (GHD-9070A, JingHong, Shanghai, China) at 80
˚
C for four minutes. After being cooled in a glass desiccator, they were imaged and color measured again. The same operation was run three more times with the same drying temperature (the third operation for six minutes, the fourthoperation for eight minutes and the fifth operation for ten minutes).
The main steps of this study are illustrated in Fig. 3. All raw hyperspectral images were acquired by the hyperspectral imaging system in the wavelengths of 380 to 1030 nm. Simultaneously, color values were determined by the
Fig. 2. Schematic diagram of the hyperspectral imaging system.
colorimeter. The raw hyperspectral images were then corrected by the dark and white reference images. Spectral reflectance values of all pixels from the ROI (30630 pixels) of each sample were extracted and averaged as one variable. Thus,
a total of 210Xvariables were obtained and used to represent the spectral data of all samples. Then, these samples were divided into two sets at a ratio of 2:1 (2/3 for calibration and 1/3 for prediction). Nine different pre-processing methods were used to improve the predictive ability. In order to optimize calibration model, two effective wavelengths selection methods including competitive adaptive reweighted sampling (CARS) and successive projections algorithm (SPA) were used to select the key wavelengths. The optimal calibration model was determined according to the values of rc, rp,RMSEC and RMSEP.
Image acquisition and correction
The exposure time was 0.07 s, the moving speed was 2.6 mm/sec, and the vertical distance between the lens and sample was 36.0 cm. Each leaf was then placed on
Fig. 3. Main steps of this study.
the conveyer belt to be scanned using the hyperspectral imaging system. Then, one hyperspectral image (hyperspectral cube) for each sample was obtained covering the spectral wavelengths of 380 to 1030 nm. The dimensions of the hyperspectral cube were 512 bands in theldimension and 672 pixels in theydimension. When
raw hyperspectral images were created, they should be corrected with dark and white reference images based on equation (1). The dark reference image with the reflectance factor of about 0% was obtained by covering the lens with the cap and turning off the light. The white reference image with the reflectance factor of about 99% was obtained from a white Teflon board (CAL-tile200, 200 mm625
mm610 mm).
Rcal~ Iraw{Idark
Iwhite{Idark ð1Þ
WhereRcal is the corrected hyperspectral image,Iraw is the raw hyperspectral
image, Idark is the dark reference image, andIwhite is the white reference image.
Models and evaluation index
during different drying periods. The calculation was carried out by the free LS-SVM toolbox (LS-LS-SVM v1.5, Suykens, Leuven, Belgium) in Matlab R2009a.
The performance of the models was evaluated in accordance with the values of the correlation coefficient (rc and rp), root mean square error of calibration (RMSEC) and root mean square error of prediction (RMSEP) [29]. A good model should be of high values ofrcand rp, low values ofRMSECandRMSEP, and small
difference between RMSECand RMSEP. A large difference between the values of
RMSEC and RMSEPindicates that the model is over-fitting [30].
Key wavelengths selection
For the purpose of improving the performance of the predictive ability and reducing the influence of redundant information between contiguous wavelengths in the whole spectrum, selection of effective wavelengths is a very significant operation in spectral studies [31]. According to the previous studies, the selected wavelengths can be equally or even more effective than the whole spectral wavelengths [32]. The CARS was firstly used to select effective wavelengths from the full spectral wavelengths in this study. The CARS selects the key wavelengths based on the principle of ‘‘survival of the fittest’’ [33]. It abandons the
wavelengths which are of small regression coefficients by exponentially decreasing function (EDF). The main procedures of each sampling run can be described as follows: (a) model sampling based on Monte Carlo (MC) principle; (b)
wavelengths selection using EDF; (c) competitive wavelengths selection based on adaptive reweighted sampling (ARS); (d) evaluation of the subset by cross validation. Finally, those wavelengths which contain little or no useful
information are eliminated while effective wavelengths are retained [34], [35]. The SPA, which is also a robust method for the selection of key wavelengths, was then used to select effective wavelengths. This algorithm can solve the collinear problem by selecting optimal wavelengths with minimal redundancy, and use a projection operation in a vector space for selecting key wavelengths [36], [37]. Both of the two wavelengths selection algorithms were operated in Matlab R2009a. Finally, the raw spectral data were consequently reduced into a matrix with a dimension of m6n(m was the number of the samples, and n was the number of the selected wavelengths).
Measurement of color values (
D
L*,
D
a* and
D
b*)
In tea processing procedure, color parameters of tea leaves play vital roles for the reason that they can not only directly determine tea’s quality, but also reflect the quality. Therefore, it is crucial to acquire the color parameters of tea leaves at different drying periods. In this study, three color parameters (DL*,Da* andDb*) were measured using the colorimeter. Before color measurement, the colorimeter should be firstly calibrated by a standard white calibration plate. The parameter
DL* is the lightness or luminance component. The other two parameters (Da* and
parameters DL*, Da* andDb* represent color changes from dark to brightness, green to red and blue to yellow, respectively.
Results and Discussion
Statistics of color parameters
A total of 210 tea leaves at five different drying periods (fresh, drying for 4 min, drying for 6 min, drying for 8 min and drying for 10 min) were studied. They were divided into the calibration set and the prediction set at a ratio of 2:1. That is one sample was picked out from every three ones consecutively, which resulted in 140 samples for the calibration set (Biyun: 46, Longjing-43: 47, Zhongcha-302: 47) and 70 ones for the prediction set (Biyun: 24, Longjing-43: 23, Zhongcha-302: 23). The detailed statistical values of each set are shown in Table 1.
Preprocessing results
In order to obtain useful spectral information and improve the predictive ability, wavelengths at the beginning with some noise were rejected, resulting in the wavelengths of 400 to 1030 nm were studied. Then, nine different preprocessing algorithms were used to evaluate the optimal one in terms of the values of rc, rp, RMSEC andRMSEP of PLS model. The results can be seen in Table 2. Based on the evaluation standards, raw data performed best with the highest values of rp
(0.925 forDL* and 0.930 forDb*, respectively). Though the PLS model based on MSC preprocessing method obtained the result with the highest values of rc
(0.949) andrp(0.799) forDa*, the lowest value ofRMSEC(0.611) and the second
lowest value of RMSEP (1.276), it did not performed well due to the big gap between the values of rcand rp. Thus, raw data was used for further study.
Effective wavelengths
In order to obtain the optimal model with the robust predictive ability and a small number of input variable, two wavelengths selection methods (CARS and SPA) were conducted to determine the most effective wavelengths in this study, respectively. As a result, forty-eight (DL*), thirty-four (Da*) and twenty-six (Db*) wavelengths were identified by CARS, respectively; seven (DL*), six (Da*) and eleven (Db*) wavelengths were selected by SPA, respectively. Compared with the number of full spectral wavebands, those of the selected wavelengths
and speed up the calculation but also improve the accuracy and robustness of the predictive ability.
Predicted results
The predicted results can be seen inTable 4. On the basis of the evaluation index, CARS-MLR model performed perfectly with a satisfying result forDL* (rc50.963, rp50.931, RMSEC50.838 and RMSEP51.160). For Da*, CARS-LS-SVM model obtained the best result with the rcof 0.968, rp of 0.842, RMSECof 0.492 and RMSEP of 1.164, respectively. ForDb*, it was also the CARS-LS-SVM model that performed excellently with the rc of 0.965,rp of 0.944, RMSEC of 1.381 and RMSEP of 1.762, respectively. The number of input variables for these three models was 48, 34 and 26, respectively. Among the four models established based on SPA, SPA-LS-SVM model performed best with the highest values of rc(0.933
for DL*, 0.893 for Da* and 0.916 for Db*, respectively) and rp (0.929 forDL*, 0.849 forDa* and 0.917 forDb*, respectively), the lowest values ofRMSEC(1.116 for DL*, 0.877 for Da* and 2.133 for Db*, respectively) and RMSEP (1.178 for
DL*, 1.146 forDa* and 2.142 forDb*, respectively). From the results, it could be seen that LS-SVM model based on the selected wavelengths performed better than other models. Though there was a little decrement of the values of rcand rp,
increment of the values ofRMSECandRMSEPfor those models established based on SPA, the input variables were fewer compared with the models which were established based on CARS, respectively. The fewer input variables demonstrated that CARS and SPA can improve the performance of the predicted ability for the determination of color parameters. Thus, these selected wavelengths were more efficient than the whole wavelengths. It may because that the whole spectral wavelengths contained more redundant information which affects the perfor-mance of the predicted results. The predicted results of CARS-LS-SVM and SPA-LS-SVM models were shown in Fig. 4 (a–f), respectively. It could be found that the plots in calibration and prediction sets were distributed near the ideal lines, indicating that the performance of these models were good. It demonstrated that hyperspectral imaging technique could be used to determine the color parameters of tea leaves, and both CARS and SPA methods could remove uninformative wavelengths and improve the predicted ability of models.
Table 1.Reference values of color (DL*,Da* andDb*) of tea leaves in calibration and prediction sets.
Statistics Calibration Prediction
DL* Da* Db* DL* Da* Db*
Minimum 269.78 29.66 7.56 269.96 29.27 8.87
Maximum 255.59 1.81 30.43 256.35 2.11 31.84
Mean 262.77 25.85 16.85 262.78 25.74 17.05
Standard Deviation 3.13 1.95 5.33 3.20 2.14 5.40
Table 2.Performance of models in calibration and prediction for predicting color (DL*,Da* andDb*) using different preprocessing methods.
Preprocessing Calibration Prediction No.h
DL* Da* Db* DL* Da* Db*
rc RMSEC rc RMSEC rc RMSEC rp RMSEP rp RMSEP rp RMSEP
Raw 0.911 1.286 0.837 1.064 0.917 2.118 0.925 1.205 0.793 1.295 0.930 1.979 5/9/8 MASa 0.910 1.288 0.898 0.855 0.918 2.107 0.925 1.204 0.781 1.327 0.928 1.991 5/13/9
SGSb 0.909 1.296 0.910 0.805 0.914 2.156 0.925 1.204 0.782 1.324 0.923 2.061 5/14/8
MFSc 0.910 1.289 0.896 0.863 0.913 2.166 0.925 1.207 0.780 1.328 0.925 2.045 5/13/8 GFSd 0.910 1.290 0.834 1.071 0.917 2.126 0.925 1.209 0.785 1.317 0.928 1.991 5/9/8 Normalize 0.912 1.276 0.820 1.113 0.917 2.117 0.925 1.209 0.793 1.294 0.927 2.011 4/7/7 MSCe 0.898 1.368 0.949 0.611 0.924 2.038 0.906 1.342 0.799 1.276 0.908 2.253 5/15/8
SGDf 0.947 1.004 0.775 1.229 0.961 1.472 0.914 1.289 0.618 1.670 0.904 2.293 7/4/8
Baseline 0.892 1.410 0.863 0.981 0.922 2.058 0.902 1.369 0.791 1.301 0.925 2.045 4/10/9 SNVg 0.898 1.367 0.942 0.652 0.924 2.037 0.906 1.342 0.804 1.263 0.909 2.236 5/14/8
a
Moving average smoothing;
bSavitzky-Golay smoothing; cMedian filter smoothing; dGaussian filter smoothing; e
Multiplicative scatter correction;
f
Savitzky-Golay derivatives;
g
Standard normal variate;
hNumber of latent variables of
DL*/Number of latent variables ofDa*/Number of latent variables ofDb*.
doi:10.1371/journal.pone.0113422.t002
Color
Measureme
ONE
|
DOI:10.13
71/journal.p
one.01
13422
December
29,
Classification of different samples
The correct classification rates (CCRs) of samples at five different drying periods based on LS-SVM model were shown inTable 5. The results covered from 89.29% to 100% in the calibration set and from 71.43% to 100% in the prediction set, respectively. The total CCRs were 96.43% in the calibration set and 85.71% in the prediction set, respectively. From the results, it can be seen that type (6 min) and type (8 min) were identified badly (both were 71.43%), other three types were classified excellently with high values of CCRs (from 92.86% to 100%). There were three samples in type (6 min) were identified as type (8 min), and four ones in type (8 min) were identified as type (6 min). This might because of the short interval of drying time, which caused tiny change of color parameters of tea leaves.
Table 3.Effective wavelengths recommended by CARS and SPA, respectively.
Methods Type Number Effected wavelengths/nm
CARS DL* 48 410, 413, 416, 420, 425, 429, 444, 448, 531, 539, 543, 544, 545, 638, 639, 677, 767, 869, 874, 884, 892, 922, 925, 929, 934, 937, 945, 958, 965, 969, 971, 973, 975, 977, 978, 981, 982, 985, 987, 998, 999, 1001, 1006, 1007, 1013, 1015, 1018, 1027
CARS Da* 34 407, 428, 429, 515, 517, 518, 519, 540, 543, 544, 548, 586, 588, 590, 610, 611, 613, 614, 615, 616, 676, 693, 724, 741, 922, 924, 950, 965, 971, 985, 986, 1014, 1017, 1021
CARS Db* 26 461, 462, 543, 545, 584, 585, 608, 609, 610, 700, 711, 729, 753, 846, 848, 965, 969, 974, 975, 977, 987, 989, 990, 1009, 1015, 1018
SPA DL* 7 457, 540, 649, 735, 761, 874, 1017 SPA Da* 6 540, 608, 676, 690, 985, 1017
SPA Db* 11 404, 408, 414, 416, 418, 444, 540, 648, 770, 866, 971
doi:10.1371/journal.pone.0113422.t003
Table 4.Performance of different models in calibration and prediction for predicting color (DL*,Da* andDb*).
Model Calibration Prediction Bandsa
DL* Da* Db* DL* Da* Db*
rc RMSEC rc RMSEC rc RMSEC rp RMSEP rp RMSEP rp RMSEP
LS-SVM 0.972 0.741 0.973 0.463 0.967 1.354 0.931 1.164 0.973 1.051 0.933 1.936 496/496/ 496 PCR 0.908 1.302 0.799 1.169 0.915 2.134 0.927 1.188 0.762 1.375 0.921 2.083 496/496/
496 CARS-LS-SVM 0.980 0.616 0.968 0.492 0.965 1.381 0.920 1.251 0.842 1.164 0.944 1.762 48/34/26 CARS-PLS 0.955 0.930 0.874 0.944 0.944 1.752 0.924 1.219 0.806 1.257 0.925 2.036 48/34/26 CARS-PCR 0.904 1.333 0.873 0.948 0.934 1.901 0.922 1.223 0.812 1.241 0.915 2.159 48/34/26 CARS-MLR 0.963 0.838 0.907 0.820 0.951 1.648 0.931 1.160 0.807 1.256 0.931 1.958 48/34/26 SPA-LS-SVM 0.933 1.116 0.893 0.877 0.916 2.133 0.929 1.178 0.849 1.146 0.917 2.142 7/6/11 SPA-PLS 0.914 1.263 0.809 1.143 0.904 2.265 0.925 1.205 0.746 1.415 0.907 2.258 7/6/11 SPA-PCR 0.911 1.285 0.669 1.446 0.891 2.417 0.924 1.219 0.754 1.397 0.904 2.287 7/6/11 SPA-MLR 0.917 1.247 0.810 1.140 0.906 2.244 0.926 1.201 0.754 1.397 0.908 2.254 7/6/11
aNumber of input bands of
DL*/Number of input bands ofDa*/Number of input bands ofDb*.
However, the total result was acceptable which demonstrated that hyperspectral imaging technique combined with LS-SVM model could also be used to classify tea leaves at different drying periods.
Fig. 4. Measured vs. predicted values of calibration and prediction by CARS-LS-SVM and SPA-LS-SVM models, respectively.(a):
CARS-LS-SVM-DL*; (b): CARS-LS-SVM-Da*; (c): CARS-LS-SVM-Db*; (d): SPA-LS-SVM-DL*; e): SPA-LS-SVM-Da*; (f): SPA-LS-SVM-Db*.
doi:10.1371/journal.pone.0113422.g004
Table 5.Correct classification rates based on LS-SVM.
Calibration set Prediction set
Types No. Missed CCRa/% No. Missed CCRa/%
0 min 28 0 100 14 1 92.86
4 min 28 0 100 14 0 100
6 min 28 2 92.86 14 4 71.43
8 min 28 3 89.29 14 4 71.43
10 min 28 0 100 14 1 92.86
Total 140 5 96.43 70 10 85.71
aCorrect classification rates.
Conclusions
This study was carried out to evaluate the feasibility of using visible and near infrared hyperspectral imaging technique, which covers the spectral wavelengths of 380 to 1030 nm, to determine color parameters of tea leaves during different drying periods. Two wavelengths selection methods including CARS and SPA were used to select effective wavelengths. Four different models (LS-SVM, PLS, PCR and MLR) were used to predict color values. Each wavelength selection method and each model obtained a good result. Among all models, the values of
rpranged from 0.902 to 0.931 forDL*, from 0.618 to 0.973 forDa* and from 0.904
to 0.944 for Db*, respectively. Based on the selected wavelengths, multispectral imaging system could be designed for nondestructive quality inspection during tea processing industry. Moreover, the CCRs of tea leaves at five different drying periods based on LS-SVM model covered from 89.29% to 100% in calibration set and from 71.43% to 100% in prediction set, respectively. The result demonstrates that this technique could to be used as an objective and nondestructive method to determine color parameters of tea leaves and classify samples during different drying periods. This is the first time that the visible and near infrared
hyperspectral imaging technique was applied in the color determination of tea leaves at different drying periods. This technique can also be considered to determine some other chemical components which are also very important for tea’s quality.
However, this study was a preliminary work. In further studies, more samples with different drying time should be selected to build more accurate and robust model. More effective wavelengths with higher accuracy and fewer variables should also be considered.
Acknowledgments
This work was supported by 863 National High-Tech Research and Development Plan (2013AA102301, 2011AA100705), Zhejiang Province Department of
education research project (Y201327409), the Fundamental Research Funds for the Central Universities of China (2012FZA6005, 2013QNA6011) and National Natural Science Foundation of China (61201073).
Author Contributions
Conceived and designed the experiments: CX YS YH. Performed the experiments: CX YH. Analyzed the data: CX YS YH. Contributed reagents/materials/analysis tools: CX XL YS YH. Wrote the paper: CX YH.
References
2. Nakachi K, Matsuyama S, Miyake S, Suganuma M, Imai K(2000) Preventive effects of drinking green tea on cancer and cardiovascular disease: Epidemiological evidence for multiple targeting prevention. BioFactors 13: 49–54.
3. Xie CQ, Li XL, Nie PC, He Y(2013) Application of time series hyperspectral imaging (TS-HSI) for determining water content within tea leaves during drying. T ASABE 56: 1431–1440.
4. Barbin DF, ElMasry G, Sun DW, Allen P(2012) Predicting quality and sensory attributes of pork using near-infrared hyperspectral imaging. Anal Chim Acta 719: 30–42.
5. Deng SG, Xu YF, Li L, Li XL, He Y (2013) A feature-selection algorithm based on support vector machine-multiclass for hyperspectral visible spectral analysis. J Food Eng 119: 159–166.
6. Wei X, Liu F, Qiu ZJ, Shao YN, He Y(2014) Ripeness classification of astringent persimmon using hyperspectral imaging technique. Food Bioprocess Tech 7: 1371–1380.
7. Wu D, Wang SJ, Wang NF, Nie PC, He Y, et al.(2013) Application of time series hyperspectral imaging (TS-HSI) for determining water distribution within beef and spectral kinetics analysis during dehydration. Food Bioprocess Tech 6: 2943–2958.
8. Wu D, Sun DW (2013) Potential of time series-hyperspectral imaging (TS-HSI) for non-invasive determination of microbial spoilage of salmon flesh. Talanta 111: 39–46.
9. Huang M, Wang QG, Zhang M, Zhu QB(2014) Precition of color and moisture content for vegetable soybean during drying using hyperspectral imaging technology. J Food Eng 128: 24–30.
10. Iqbal A, Sun DW, Allen P(2013) Predciting of moisture, color and pH in cooked, pre-sliced turkey hams by NIR hyperspectral imaging system. J Food Eng 117: 42–51.
11. Karimi Y, Maftoonazad N, Ramaswamy HS, Prasher SO, Marcotte M (2012) Application of hyperspectral technique for color classification avocados subjected to different treatments. Food Bioprocess Tech 5: 252–264.
12. Wu JH, Peng YK, Li YY, Chen JJ, Dhakal S(2012) Prediction of beer quality attributes using VIS/NIR hyperspectral scattering imaging technique. J Food Eng 109: 267–273.
13. Wu D, Sun DW, He Y (2012) Application of long-wave near infrared hyperspectral imaging for measurement of color distribution in salmon fillet. Innov Food Sci Emerg 16: 361–372.
14. Cen HY, Bao YD, He Y, Sun DW(2007) Visible and nearinfraredspectroscopy for rapiddetection of citric and tartaricacids in orange juice. J Food Eng 82: 253–260.
15. He Y, Huang M, Garcia A, Hernandez A, Song HY(2007) Prediction of soil macronutrients content using near-infrared spectroscopy. Comput Electron Agr 58: 144–153.
16. Jiang LL, Liu F, He Y (2012) A non-destructive distinctive method for discrimination of automobile lubricant variety by visible and short-wave infrared spectroscopy. Sensors 12: 3498–3511.
17. Shao YN, He Y, Go´mez AH, Pereir AG, Qiu ZJ, et al. (2007) Visible/near infrared spectrometric technique for nondestructive assessment of tomato ‘Heatwave’ (Lycopersicumesculentum) quality characteristics. J Food Eng 81: 672–678.
18. Zou XB, Shi JY, Hao LM, Zhao JW, Mao HP, et al.(2011) In vivo noninvasive detection of chlorophyll distribution in cucumber (Cucumissativus) leaves by indices based on hyperspectral imaging. Anal Chim Acta 706: 105–112.
19. Zhu FL, Cheng SX, Wu D, He Y(2011) Rapid discrimination of fish feeds brands based on visible and short-wave near-infrared spectroscopy. Food Bioprocess Tech 4: 597–602.
20. Kamruzzaman M, ElMasry G, Sun DW, Allen P(2011) Prediction of some quality attributes of lamb meat using near-infrared hyperspectral imaging and multivariate analysis. Anal Chim Acta 714: 57–67.
21. Shao XG, Wang W, Hou ZY, Cai WS (2006) A new regression method based on independent component analysis. Talanta 69: 676–680.
22. Wang W, Peng YK, Huang H, Wu JH(2011) Application of Hyper-Spectral imaging technique for the detection of total viable bacteria count in pork. Sens Lett 9: 1024–1030.
23. Bieroza M, Baker A, Bridgeman J (2012) New data mining and calibration approaches to the assessment of water treatment efficiency. Adv Eng Softw 44: 126–135.
25. Næs T, Mevik BH(2001) Understanding the collinearity problem in regression and discriminant analysis. J Chemometr 15: 413–426.
26. Suykens JAK, Vandewalle J(1999) Least squares support vector machine classifiers. Neural Process Lett 9: 293–300.
27. Chen XJ, Wu D, He Y, Liu S(2011) Nondestructive differentiation of panax species using visible and shortwave near-infrared spectroscopy. Food Bioprocess Tech 4: 753–761.
28. Zhang XL, Liu F, He Y, Li XL(2012) Application of hyperspectral imaging and chemometric calibration for variety discrimination of maize seeds. Sensors 12: 17234–17246.
29. Wu D, Sun DW(2013) Application of visible and near infrared hyperspectral imaging for non-invasively measuring distribution of water-holding capacity in salmon flesh. Talanta 116: 266–276.
30. Wu D, Chen XJ, Zhu XG, Guan XC, Wu GC(2011) Uninformative variable elimination for improvement of successive projections algorithm on spectral multivariable selection with different calibration algorithms for the rapid and non-destructive determination of protein content in dried laver. Anal Methods 3: 1790–1796.
31. ElMasry G, Sun DW, Allen P(2012) Near-infrared hyperspectral imaging for predicting colour, pH and tenderness of fresh beef. J Food Eng 110: 127–140.
32. Kamruzzaman M, ElMasry G, Sun DW, Allen P(2011) Application of NIR hyperspectral imaging for discrimination of lamb muscles. J Food Eng 104: 332–340.
33. Li HD, Liang YZ, Xu QS, Cao DS (2009) Key wavelength screening using competitive adaptive reweighted sampling method for multivariate calibration. Anal Chim Acta 648: 77–84.
34. Wu D, Sun DW (2013) Potential of time series-hyperspectral imaging (TS-HSI) for non-invasive determination of microbial spoilage of salmon flesh. Talanta 111: 39–46.
35. Wei X, Xu N, Wu D, He Y(2014) Determination of branched-amino acid content in fermented cordyces sinensis mycelium by using FT-NIR spectroscopy technique. Food Bioprocess Tech 7: 184–190.
36. Arau´jo MCU, Saldanha TCB, Galva˜o RKH, Yoneyama T, Chame HC, et al.(2001) The successive projections algorithm for variable selection in spectroscopicmulticomponent analysis. Chemometr Intell Lab 57: 65–73.
